We propose experiments designed to increase our understanding of the roles for certain types of cell adhesion molecules in learning and memory. Prior research has indicated that certain members of the integrin family of cell adhesion molecules, and members of the immunoglobulin superfamily of cell adhesion molecules, are important for odor learning in Drosophila. Mutants that are defective in the expression of these molecules will be studied molecularly and behaviorally. Some experiments will determine whether expression of transgenes in restricted portions of the brain will rescue the mutant learning deficits. This will reveal whether limited spatial expression is sufficient for behavioral rescue. The mutants will also be tested in operant learning situations to determine whether the behavioral deficit is specific for odor learning, or whether the molecules have a more general influence on conditioned behavior. Over-expression of some cell adhesion molecules blocks long-term memory. Experiments are proposed to define the time window of over-expression necessary to observe the block. And, we will pursue an understanding of how these cell adhesion molecules mediate behavior by studying molecules known to organize adhesion molecules at the synapse, and candidate signaling molecules known to function downstream of the cell adhesion receptors. These experiments will further our understanding of the molecular and cellular basis of learning and memory in Drosophila. This basic information is expected to be of general significance to learning in other species and of relevance to diseases that affect cognition.